US5663763AExpiredUtility
Picture signal encoding method and apparatus and picture signal decoding method and apparatus
Est. expiryOct 29, 2012(expired)· nominal 20-yr term from priority
H04N 19/61H04N 19/91H04N 19/577H04N 19/126H04N 19/13H04N 19/124H04N 19/60H04N 19/85H04N 19/112
84
PatentIndex Score
62
Cited by
24
References
13
Claims
Abstract
An encoding and decoding apparatus and method which has the primary benefit over the prior art of not requiring storage of the nonlinear sequence of numbers representative of the quantization characteristics. The hardware required for quantization is reduced and inverse quantization may be achieved using shifting operations and a multiplier of up to half the scale compared to conventional encoding/decoding methods and encoding/decoding apparatus.
Claims
exact text as granted — not AI-modifiedWe claim:
1. An encoding method for picture signals in which input picture signals are quantized and subsequently encoded, comprising setting a digital signal value representing a quantization information used for expressing an index of powers of 2 as a first quantization information among a set of quantization information, setting a digital signal value corresponding to a coefficient multiplied by the powers of 2 as representing a second quantization information among the set of quantization information, and quantizing the picture signals based on non-linear quantization characteristic signals (QUANT) obtained by using k as said first quantization information, k being a positive integer, and by using (i/2+j) as said second quantization information, j being a positive integer and i being 0 or 1, wherein said non-linear quantization characteristic signals (QUANT) are represented by an equation: QUANT=(i/2+j)×2.sup.k +2.sup.(k+2) -4 and used as a width of quantization to produce quantized picture signals.
2. The picture signal encoding method as claimed in claim 1 wherein said set of quantization information comprising said first quantization information and said second quantization information is represented by five bits.
3. The picture signal encoding method as claimed in claim 2 wherein the relation between k, i and j and the non-linear quantization characteristic signals are as shown in the following Table 1: TABLE 1
______________________________________
quantization
quantization characteristics
index information decimal binary
number k j i expression
expression
______________________________________
0 00 00 0 inhibit
inhibit
1 00 00 1 0.5 00000.1
2 00 01 0 1.0 00001.0
3 00 01 1 1.5 00001.1
4 00 10 0 2.0 00010.0
5 00 10 1 2.5 00010.1
6 00 11 0 3.0 00011.0
7 00 11 1 3.5 00011.1
8 01 00 0 4.0 000100.
9 01 00 1 5.0 000101.
10 01 01 0 6.0 000110.
11 01 01 1 7.0 000111.
12 01 10 0 8.0 001000.
13 01 10 1 9.0 001001.
14 01 11 0 10.0 001010.
15 01 11 1 11.0 001011.
16 10 00 0 12.0 001100.
17 10 00 1 14.0 001110.
18 10 01 0 16.0 010000.
19 10 01 1 18.0 010010.
20 10 10 0 20.0 010100.
21 10 10 1 22.0 010110.
22 10 11 0 24.0 011000.
23 10 11 1 26.0 011010.
24 11 00 0 28.0 011100.
25 11 00 1 32.0 100000.
26 11 01 0 36.0 100100.
27 11 01 1 40.0 101000.
28 11 10 0 44.0 101100.
29 11 10 1 48.0 110000.
30 11 11 0 52.0 110100.
31 11 11 1 56.0 111000.
______________________________________
4. The picture signal encoding method as claimed in claim 3 wherein, if said non-linear quantization characteristic signals (OUANT) are represented by a binary number, there exists at least one effective bit in four consecutive bits.
5. The picture signal encoding method as claimed in claim 2 wherein the relation between k, i, and j and the non-linear quantization characteristic signals are as shown in the following Table 2: TABLE 2
______________________________________
quantization
quantization characteristics
index information decimal binary
number k j i expression
expression
______________________________________
0 00 00 0 inhibit
inhibit
1 00 00 1 0.5 00000.1
2 00 01 0 1.0 00001.0
3 00 01 1 1.5 00001.1
4 00 10 0 2.0 00010.0
5 00 10 1 2.5 00010.1
6 00 11 0 3.0 00011.0
7 00 11 1 3.5 00011.1
8 01 00 0 4.0 000100.
9 01 00 1 5.0 000101.
10 01 01 0 6.0 000110.
11 01 01 1 7.0 000111.
12 01 10 0 8.0 001000.
13 01 10 1 9.0 001001.
14 01 11 0 10.0 001010.
15 01 11 1 11.0 001011.
16 10 00 0 12.0 001100.
17 10 00 1 14.0 001110.
18 10 01 0 16.0 010000.
19 10 01 1 18.0 010010.
20 10 10 0 20.0 010100.
21 10 10 1 22.0 010110.
22 10 11 0 24.0 011000.
23 10 11 1 26.0 011010.
24 11 00 0 28.0 011100.
25 11 00 1 36.0 100100.
26 11 01 0 44.0 101100.
27 11 01 1 52.0 110100.
28 11 10 0 60.0 111100.
29 11 10 1 68.0 1000100.
30 11 11 0 76.0 1001100.
31 11 11 1 84.0 1010100.
______________________________________
6. The picture signal encoding method as claimed in claim 5 wherein there exists at least one effective bit in four consecutive bits.
7. A picture signal decoding method in which transmitted encoded data is inverse-quantized and subsequently decoded for restoring a picture from the encoded data, comprising the steps of reproducing non-linear quantization characteristic signals (QUANT) at the time of inverse quantization by multiplying powers of 2 by a coefficient, using a digital signal value for expressing an index of the powers of 2 as representing a first quantization information among a set of quantization information and a digital signal value corresponding to said coefficient as representing a second quantization information, and inverse guantizing data based on the reproduced non-linear quantization characteristic signals (QUANT) obtained by using k as said first quantization information and by using (i/2+j) as said second quantization information, j and k being positive integers and i being 0 or 1, wherein said non-linear quantization characteristic signals (QUANT) are represented by an equation: QUANT=(i/2+j)×2.sup.k +2.sup.(k+2) -4 and used as a width of quantization to reproduce quantized picture signals.
8. The picture signal decoding method as claimed in claim 7 wherein said set of quantization information comprising said first quantization information and said second quantization information is represented by five bits.
9. The picture signal decoding method as claimed in claim 8 wherein the relation between k, i and j and the non-linear quantization characteristic signals are as shown in the following Table 5: TABLE 5
______________________________________
quantization
quantization characteristics
index information decimal binary
number k j i expression
expression
______________________________________
0 00 00 0 inhibit
inhibit
1 00 00 1 0.5 00000.1
2 00 01 0 1.0 00001.0
3 00 01 1 1.5 00001.1
4 00 10 0 2.0 00010.0
5 00 10 1 2.5 00010.1
6 00 11 0 3.0 00011.0
7 00 11 1 3.5 00011.1
8 01 00 0 4.0 000100.
9 01 00 1 5.0 000101.
10 01 01 0 6.0 000110.
11 01 01 1 7.0 000111.
12 01 10 0 8.0 001000.
13 01 10 1 9.0 001001.
14 01 11 0 10.0 001010.
15 01 11 1 11.0 001011.
16 10 00 0 12.0 001100.
17 10 00 1 14.0 001110.
18 10 01 0 16.0 010000.
19 10 01 1 18.0 010010.
20 10 10 0 20.0 010100.
21 10 10 1 22.0 010110.
22 10 11 0 24.0 011000.
23 10 11 1 26.0 011010.
24 11 00 0 28.0 011100.
25 11 00 1 36.0 100100.
26 11 01 0 44.0 101100.
27 11 01 1 52.0 110100.
28 11 10 0 60.0 111100.
29 11 10 1 68.0 1000100.
30 11 11 0 76.0 1001100.
31 11 11 1 84.0 1010100.
______________________________________
10. The picture signal decoding method as claimed in claim 8 wherein, in inverse-quantizing the encoded data, the encoded data is multiplied in a three stage adder and a resulting sum is shifted by a number of bits determined by said first quantization information.
11. The picture signal decoding method as claimed in claim 8 wherein the relation between information k, i and j and the non-linear quantization characteristic signals are set as shown in the following Table 4: TABLE 4
______________________________________
quantization
quantization characteristics
index information decimal binary
number k j i expression
expression
______________________________________
0 00 00 0 inhibit inhibit
1 00 00 1 0.5 00000.1
2 00 01 0 1.0 00001.0
3 00 01 1 1.5 00001.1
4 00 10 0 2.0 00010.0
5 00 10 1 2.5 00010.1
6 00 11 0 3.0 00011.0
7 00 11 1 3.5 00011.1
8 01 00 0 4.0 000100.
9 01 00 1 5.0 000101.
10 01 01 0 6.0 000110.
11 01 01 1 7.0 000111.
12 01 10 0 8.0 001000.
13 01 10 1 9.0 001001.
14 01 11 0 10.0 001010.
15 01 11 1 11.0 001011.
16 10 00 0 12.0 001100.
17 10 00 1 14.0 001110.
18 10 01 0 16.0 010000.
19 10 01 1 18.0 010010.
20 10 10 0 20.0 010100.
21 10 10 1 22.0 010110.
22 10 11 0 24.0 011000.
23 10 11 1 26.0 011010.
24 11 00 0 28.0 011100.
25 11 00 1 32.0 100000.
26 11 01 0 36.0 100100.
27 11 01 1 40.0 101000.
28 11 10 0 44.0 101100.
29 11 10 1 48.0 110000.
30 11 11 0 52.0 110100.
31 11 11 1 56.0 111000.
______________________________________
12. An encoding apparatus for encoding picture signals in which input picture signals are quantized and subsequently encoded, comprising: means for setting a digital signal value representing a quantization information used for expressing an index of powers of 2 as a first quantization information among a set of quantization information, means for setting a digital signal value corresponding to a coefficient multiplied by the powers of 2 as representing a second quantization information among the set of quantization information, and means for quantizing the picture signals based on non-linear quantization characteristic signals (QUANT) obtained by using k as said first quantization information, k being a positive integer, and by using (i/2+j) as said second quantization information, j being a positive integer and i being 0 or 1, wherein said non-linear quantization characteristic signals (QUANT) are represented by an equation: QUANT=(i/2+j)×2.sup.k +2.sup.(k+2) -4 and used as a width of quantization to produce quantized picture signals.
13. A picture signal decoding apparatus in which transmitted encoded data is inverse-quantized and subsequently decoded for restoring a picture from coded data, comprising: means for reproducing non-linear quantization characteristic signals (QUANT) at the time of inverse quantization by multiplying powers of 2 by a coefficient, using a digital signal value for expressing an index of the powers of 2 as representing a first quantization information among a set of quantization information and a digital signal value corresponding to said coefficient as representing a second quantization information, and means for inverse quantizing data based on the reproduced non-linear quantization characteristic signals (QUANT) obtained by using k as said first quantization information and by using (i/2+j) as said second quantization information, j and k being positive integers and i being 0 or 1, wherein said non-linear quantization characteristic signals (QUANT) are represented by an equation: QUANT=(i/2+j)×2.sup.k +2.sup.(k+2) -4 and used as a width of quantization to reproduce quantized picture signals.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.